Senescence, a permanent form of growth arrest following oncogene activation or telomere attrition, is generally considered a tumor-suppressive mechanism active in vitro and in vivo. However, senescent fibroblasts that are near epithelial cancer cells may promote tumor formation, although in vivo evidence for the existence of such cells and how they act to promote tumor formation remains elusive. Also unknown is how cancer cells communicate with tumor-promoting fibroblasts, if such fibroblasts do exist in vivo. We may have found a novel means for this communication: in the course of profiling RAS-transformed ovarian surface epithelial cells and their isogenic immortalized counterparts, we identified a chemokine, Gro-1, that is up-regulated in RAS- transformed ovarian cancer cell lines and is critical for transformation of ovarian epithelial cells. Unexpectedly, we found that Gro-1 induced senescence in ovarian stromal fibroblasts. Knockdown of the receptor for Gro-1, CXCR2, abrogates the senescence and leads to uncontrolled proliferation of the fibroblasts. We further demonstrated that Gro-1-induced senescent fibroblasts have an increased proangiogenic factor vascular endothelial growth factor (VEGF-A) and decreased antiangiogenic factor thrombospondin-1 (TSP-1). The ratio of VEGF-A:TSP-1 in senescent fibroblasts is 80 fold higher than that in control fibroblasts, suggesting that senescent fibroblasts provide critically needed factors to enhance tumor angiogenesis. We also observed, in the human ovarian cancer specimens, that the stromal fibroblasts near epithelial ovarian cancer cells are senescent. Because Gro-1 is a secreted molecule activated by RAS and can diffuse from epithelial cancer cells to neighboring fibroblasts, Gro-1 may be a signaling molecule by which cancer cells use to accelerate the senescence of neighboring fibroblasts. Our central hypothesis is that RAS activates Gro-1 expression in ovarian tumor cells. Gro-1 signaling through its receptor CXCR2 activates multiple downstream effectors to create a senescent phenotype. Senescent fibroblasts upregulate their VEGF-A:TSP-1 ratio to act on endothelial cells to induce an angiogenic switch, which in turn leads to tumor promotion. We propose the following two specific aims: Specific Aim 1. Define the mechanisms by which Gro-1 induces senescence and creates an angiogenic phenotype of senescent fibroblasts. Specific Aim 2. Determine the mechanisms by which senescent fibroblasts promote the tumor growth. PUBLIC HEALTH RELEVANCE: Development of cancer requires not only genetic alterations in epithelial cells but also changes in the stroma, a heterogeneous group of cells interacting with cancer cells. The predominant component of the stroma is fibroblasts. We have found that the RAS oncogene can send a signaling molecule, Gro-1, a small secreted protein, to fibroblasts to accelerate their aging process (senescence) and that this process promotes tumor formation. This project aims to define the molecular mechanisms by which aged fibroblasts initiate ovarian tumor growth and the signaling involved in the aging process. The improved understanding of the molecular mechanisms will be required for an eventual exploration of the therapeutic relevance of this observation.